Functional and evolutionary genomics of Mycobacterium tuberculosis: insights from genomic deletions in 100 strains

Abstract

To better understand genome function and evolution in Mycobacterium tuberculosis, the genomes of 100 epidemiologically well characterized clinical isolates were interrogated by DNA microarrays and sequencing. We identified 68 different large-sequence polymorphisms (comprising 186,137 bp, or 4.2% of the genome) that are present in H37Rv, but absent from one or more clinical isolates. A total of 224 genes (5.5%), including genes in all major functional categories, were found to be partially or completely deleted. Deletions are not distributed randomly throughout the genome but instead tend to be aggregated. The distinct deletions in some aggregations appear in closely related isolates, suggesting a genomically disruptive process specific to an individual mycobacterial lineage. Other genomic aggregations include distinct deletions that appear in phylogenetically unrelated isolates, suggesting that a genomic region is vulnerable throughout the species. Although the deletions identified here are evidently inessential to the causation of disease (they are found in active clinical cases), their frequency spectrum suggests that most are weakly deleterious to the pathogen. For some deletions, short-term evolutionary pressure due to the host immune system or antibiotics may favor the elimination of genes, whereas longer-term physiological requirements maintain the genes in the population.

Fig. 1.

Overall distribution of deleted sequences among 100 clinical isolates of M. tuberculosis. Sequences present in H37Rv and absent from the interrogated isolate are shown in blue. Each row represents an isolate, and each column is a region of difference. Columns are organized by genomic address; rows are organized according to phylogenetic relationships (19).

Fig. 2.

Significance of deletion proximity. Stars indicate the probability that random placement on the genome of 68 deletions of the observed sizes results in at least the observed number of pairs of deletions separated by <w bp. Diamonds indicate the ratio of expected to observed proximate pairs, n[w]_exp/n[w]_obs.

Fig. 3.

The frequency spectrum of genomic deletions. Deletion frequency (occurrences out of 100 isolates) is shown on the x axis. The number of deletions exhibiting each frequency is shown on the y axis. Phage-associated deletion 149 is not shown; it was present in 54 of 100 isolates.

Fig. 4.

Distribution of deleted genes by functional category. Number of deleted genes (blue) and nondeleted genes (yellow) are shown for each category. An asterisk indicates that a functional category was statistically overrepresented among deletions after controlling for FDR. Functional categories: 0, virulence, detoxification, adaptation; 1, lipid metabolism; 2, information pathways; 3, cell wall and cell processes; 4, stable RNAs; 5, insertion sequences and phages; 6, PE/PPE (not studied); 7, intermediary metabolism and respiration; 8, unknown; 9, regulatory proteins; 10, conserved hypotheticals.